Leptin levels in cord blood and anthropometric measures at birth: a systematic review and meta-analysis Polyxeni Karakosta a , Leda Chatzi a , Estel Plana e,f,g , Andrew Margioris b , Elias Castanas c and Manolis Kogevinas d,e,f,g Departments of a Social Medicine, b Clinical Chemistry-Biochemistry, and c Experimental Endocrinology, Faculty of Medicine, University of Crete, Heraklion, and d National School of Public Health, Athens, Greece; e Centre for Research in Environmental Epidemiology (CREAL), f Municipal Institute of Medical Research (IMIM), and g CIBER, Epidemiologia y Salud Publica, Barcelona, Spain Summary Correspondence: Dr Leda Chatzi, Department of Social Medicine, Faculty of Medicine, University of Crete, PO Box 2208, Heraklion, 71003, Crete, Greece. E-mail: lchatzi@med.uoc.gr Karakosta P, Chatzi L, Plana E, MargiorisA, Castanas E, Kogevinas M. Leptin levels in cord blood and anthropometric measures at birth: a systematic review and meta- analysis. Paediatric and Perinatal Epidemiology 2011; 25: 150–163. The role of intrauterine environment in the development of obesity is increasingly recognised. Adipokines and specifically leptin have been examined as potential biom- arkers predicting early development of obesity. We conducted a systematic review and meta-analysis of the epidemiological evidence for the association between leptin levels in cord blood and anthropometric measurements at birth in healthy mother-newborn pairs.A PubMed search was performed between 1994 and 2009 and manual search of reference lists of retrieved articles. Forty-four studies met the inclusion criteria set. All studies reported a positive correlation between leptin levels and birthweight. The combined correlation coefficient (r) was 0.46 [95%CI 0.43, 0.50]. Leptin levels explained 21% of variation in birthweight. Results were similar in males (r = 0.55; 0.40, 0.68) and females (r = 0.60; 0.50, 0.69), and between Caucasians (r = 0.45; 0.39, 0.51) and eastern Asian populations (r = 0.47; 0.37, 0.55). Statistically significant positive correlations were also found for birth length (r = 0.29; 0.23, 0.34) and ponderal index (r = 0.36; 0.31, 0.41). There was no indication of publication bias (Egger’s test P-value = 0.23). This meta-analysis shows a clear but moderate correlation between leptin levels in cord blood and birthweight that is observed in different population groups. Keywords: systematic review, meta-analysis, cord blood leptin, birthweight, birth length, birth ponderal index. Introduction During the last decades, obesity and its related conse- quences have developed into a major public health issue worldwide. The prevalence of obesity in child- hood reaches 17% among children in the USA 1 and similar percentages have been reported in other coun- tries. 2,3 A substantial body of epidemiological evidence now suggests that an adverse intrauterine environ- ment, elicited by maternal dietary or placental insuffi- ciency, may ‘programme’ susceptibility of the fetus to later development of cardiovascular or metabolic dis- eases such as obesity, hypertension, insulin resistance and type 2 diabetes. 4,5 In this regard, the study of the environment in utero has generated an effort to identify biomarkers that could predict the risk of early devel- opment of obesity. Adipokines and particularly leptin are among the most promising markers examined. Leptin, the product of the ob gene, is a proteohor- mone, which is mainly produced by adipose tissue (white and brown) and to a lesser extent by skeletal muscle, liver and placenta. 6 Shortly after its discovery, it was shown that leptin acts on hypothalamic neurons to regulate overall metabolism by decreasing the sen- sation of hunger and increasing energy expenditure. 7 Multiple subsequent studies over the last 15 years have convincingly shown that, apart from regulating satiety and energy homeostasis, leptin also exerts pleiotropic peripheral effects, as a result of the presence of leptin receptors on various organs and tissues. These actions 150 doi: 10.1111/j.1365-3016.2010.01163.x Paediatric and Perinatal Epidemiology, 25, 150–163. © 2010 Blackwell Publishing Ltd.